python wrapper for etcpak

Navigation

Verified details

These details have been verified by PyPI
Maintainers
Avatar for K0lb3 from gravatar.com K0lb3

Unverified details

These details have not been verified by PyPI
Project links
Meta
  • License: MIT License (MIT License Copyright (c) 2021 K0lb3 Permission is hereby granted, free of charge, to any person o...)
  • Author: Rudolf Kolbe
  • Tags astc, texture
  • Requires: Python >=3.7
  • Provides-Extra: tests
Classifiers

Project description

etcpak

PyPI supported Python versions Win/Mac/Linux MIT [Build Status](https://github.com/K0lb3/etcpak/actions?query=workflow Build & Publish wheels)

A python wrapper for wolfpld/etcpak All relevant function and class documentation was taken from wolfpld/etcpak.

  1. Installation
  2. Example
  3. Functions

Installation

pip install etcpak

or download/clone the git and use

python setup.py install

Example

from PIL import Image
import etcpak

# load image
img = Image.open(file_path)

# get image data
img_data = img.convert("RGBA").tobytes()

# compress data
compressed = etcpak.compress_bc3(img_data, img.width, img.height)

composite image for format comparission

import os
import etcpak
import texture2ddecoder
from PIL import Image

FORMATS = [
    ("DXT1", etcpak.compress_bc1, texture2ddecoder.decode_bc1),
    ("DXT1 Dither", etcpak.compress_bc1_dither, texture2ddecoder.decode_bc1),
    ("DXT5", etcpak.compress_bc3, texture2ddecoder.decode_bc3),
    ("ETC1", etcpak.compress_etc1_rgb, texture2ddecoder.decode_etc1),
    ("ETC1 Dither", etcpak.compress_etc1_rgb_dither, texture2ddecoder.decode_etc1),
    ("ETC2 RGB", etcpak.compress_etc2_rgb, texture2ddecoder.decode_etc2),
    ("ETC2 RGBA", etcpak.compress_etc2_rgba, texture2ddecoder.decode_etc2a8)
]

p = "S:\\Pictures"
for fp in os.listdir(p):
    if not fp[-4:] in [".png", ".jpg", ".bmp", "jpeg"]:
        continue
    # load image and adjust format and size
    print(fp)
    img = Image.open(os.path.join(p, fp)).convert("RGBA")
    img = img.crop((0,0,img.width-img.width%4, img.height-img.height%4))
    
    # create composite image
    comp = Image.new("RGBA", (img.width*8, img.height))
    comp.paste(img, (0, 0))
    print(img.width * img.height * 4)

    # iterate over all formats
    for i, (name, enc, dec) in enumerate(FORMATS):
        print(name)
        # make sure that the channel order is correct for the compression
        if name[:3] == "DXT":
            raw = img.tobytes()
        elif name[:3] == "ETC":
            r,g,b,a = img.split()
            raw = Image.merge('RGBA', (b,g,r,a)).tobytes()
        
        # compress
        data = enc(raw, img.width, img.height)

        # decompress
        dimg = Image.frombytes("RGBA", img.size, dec(data, img.width, img.height), "raw", "BGRA")

        # add to composite image
        comp.paste(dimg, (img.width*(i+1), 0))

    # save composite image
    comp.save(os.path.splitext(fp)[0]+".png")

Functions

  • all functions accept only arguments, no keywords
  • the data has to be RGBA/BGRA for the RGB functions as well
  • all DXT compressions require data in the RGBA format
  • all ETC compressions require data in the BGRA format

see etcpak/init.pyi

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page